This project will use whole-genome sequencing to answer a critical question with fundamental relevance to all neuropsychiatric diseases and possibly with fundamental relevance to development and functioning of the human brain in general: do genomic mobile element transpositions occur in the genome of brain tissue? They have been observed in cancer cells1 and are expressed in somatic tissues2. Gage's group demonstrated that L1s could retrotranspose in human neural progenitor and embryonic stem cells, and also presented PCR- based evidence suggesting an increased total number of L1 sequences in human brain regions.3 There has been no direct, sequencing-based demonstration of confirmed ME insertions in brain that are absent in other tissues. About 40% to 50% of the human genome consists of repetitive sequences known as mobile elements (MEs)4, 5, with ~33% consisting of retrotransposable elements (LINE-1, Alu, SVA). 6 These sequences (hundreds or thousands of base pairs long) are remnants of cellular or inactivated retroviral sequences which, alone or in cooperation with each other, can be transcribed and then reverse transcribed and inserted in a different location. This usually occurs in germ cells (~1 in 20 live births according to new 1000 Genomes data7). There are ~ 8,000 known polymorphic sites which are transmitted like other polymorphisms and appear subject to selection. Germline ME insertions can exert pathogenic effects by numerous mechanisms. Most aspects of normal and pathogenic ME functions remain unknown. If such genomic mobile element transposition events do occur in brain, then intensive study (of much larger brain tissue collections) will be needed to determine their functional and pathogenic effects. Any substantial increase in L1 or other MEs in brain would suggest a positively-selected functional role during normal brain development or for normal brain function (given the elegant ME inhibitory mechanisms which exist)1, 8, 9, with pathogenic defects likely to exist. Or, there could be rare (abnormal) pathogenic retroposition events (much like rare germline CNVs). But if no functionally relevant somatic cell genomic retrotransposition occurs in brain (e.g., only rare intergenic events are observed), then the role of MEs in disease can be confidently pursued with large-scale studies of genomic DNA obtained from non-brain tissue (i.e. mostly from blood or cheek swabs). Thus the answer to this question could have dramatic effects on the course of research into neural development and neuropsychiatric disease. We will include subjects with histories of schizophrenia as an example of a disease in which structural variants are known to have substantial pathogenic effects, although confirmation of the hypothesis does not depend on finding case-control differences in this study. We therefore propose to carry out whole-genome sequencing (Illumina HiSeq2000, 100bp paired-end reads, 400-500bp fragment lengths) of post-mortem brain vs. liver DNA from the same individuals to determine whether there are MEs (validated by PCR) in brain which are absent in liver, suggesting somatic cell transposition events. We will examine alternative hypotheses of uniform differences between tissues (suggesting early embryological events) vs. mosaicism within tissues (suggesting later events). The study design will utilize a combination of high- and medium-coverage sequencing of DNA from 50 individuals with schizophrenia and 50 control individuals for whom both tissues are available from the Stanley Medical Research Institute. A comprehensive pipeline for computational detection of MEs (developed in the 1000 Genomes Project by our consultant, Dr. Stewart7, and currently being benchmarked for use and installation at the Stanford Center for Genomics and Personalized Medicine where this study will be carried out) will use information from paired-end differences (one end in a unique mappable region and the other end representing repetitive sequence from a ME database) and from split reads (unique and ME sequence within a single fragment). Secondary analyses will consider issues such as the relationship between ME insertions and other structural variants, and effects of MEs on gene expression (utilizing brain expression microarray data available for some Stanley subjects). We will focus initially on superior temporal gyrus (forebrain tissue that is available in quantity). Later in the study we will study cerebellar (hindbrain) tissue from the same subjects to evaluate possible differences between brain regions.